4 code implementations • 5 Jun 2013 • Jonas Peters, Peter Bühlmann
To quantify such differences, we propose a (pre-) distance between DAGs, the structural intervention distance (SID).
2 code implementations • 10 May 2022 • Malte Londschien, Peter Bühlmann, Solt Kovács
However, the method can be paired with any classifier that yields class probability predictions, which we illustrate by also using a k-nearest neighbor classifier.
1 code implementation • 29 Oct 2020 • Yuansi Chen, Peter Bühlmann
Domain adaptation (DA) arises as an important problem in statistical machine learning when the source data used to train a model is different from the target data used to test the model.
1 code implementation • 24 Mar 2023 • Michael J. Zellinger, Peter Bühlmann
We present repliclust (from repli-cate and clust-er), a Python package for generating synthetic data sets with clusters.
3 code implementations • 27 Nov 2022 • Juan L. Gamella, Armeen Taeb, Christina Heinze-Deml, Peter Bühlmann
We leverage this procedure and evaluate the performance of GnIES on synthetic, real, and semi-synthetic data sets.
3 code implementations • 4 Jun 2018 • Niklas Pfister, Sebastian Weichwald, Peter Bühlmann, Bernhard Schölkopf
We introduce coroICA, confounding-robust independent component analysis, a novel ICA algorithm which decomposes linearly mixed multivariate observations into independent components that are corrupted (and rendered dependent) by hidden group-wise stationary confounding.
2 code implementations • 18 Jan 2018 • Dominik Rothenhäusler, Nicolai Meinshausen, Peter Bühlmann, Jonas Peters
If anchor regression and least squares provide the same answer (anchor stability), we establish that OLS parameters are invariant under certain distributional changes.
Methodology
1 code implementation • 13 Oct 2022 • Alexander Immer, Christoph Schultheiss, Julia E. Vogt, Bernhard Schölkopf, Peter Bühlmann, Alexander Marx
We study the class of location-scale or heteroscedastic noise models (LSNMs), in which the effect $Y$ can be written as a function of the cause $X$ and a noise source $N$ independent of $X$, which may be scaled by a positive function $g$ over the cause, i. e., $Y = f(X) + g(X)N$.
1 code implementation • 11 Jul 2019 • Malte Londschien, Solt Kovács, Peter Bühlmann
We propose estimation methods for change points in high-dimensional covariance structures with an emphasis on challenging scenarios with missing values.
2 code implementations • 9 Aug 2019 • Jana Janková, Rajen D. Shah, Peter Bühlmann, Richard J. Samworth
We propose a family of tests to assess the goodness-of-fit of a high-dimensional generalized linear model.
Methodology Statistics Theory Statistics Theory
1 code implementation • 5 Nov 2019 • Niklas Pfister, Evan G. Williams, Jonas Peters, Ruedi Aebersold, Peter Bühlmann
In particular, it is useful to distinguish between stable and unstable predictors, i. e., predictors which have a fixed or a changing functional dependence on the response, respectively.
Methodology Applications
1 code implementation • 8 Apr 2020 • Zijian Guo, Domagoj Ćevid, Peter Bühlmann
Inferring causal relationships or related associations from observational data can be invalidated by the existence of hidden confounding.
Methodology Statistics Theory Statistics Theory
1 code implementation • 19 Aug 2021 • Martin Emil Jakobsen, Rajen D. Shah, Peter Bühlmann, Jonas Peters
Furthermore, we study the identifiability gap, which quantifies how much better the true causal model fits the observational distribution, and prove that it is lower bounded by local properties of the causal model.
1 code implementation • 20 Jan 2021 • Lucas Kook, Beate Sick, Peter Bühlmann
In a causally inspired perspective on OOD generalization, the test data arise from a specific class of interventions on exogenous random variables of the DGP, called anchors.
Methodology
1 code implementation • 29 Jan 2021 • Corinne Emmenegger, Peter Bühlmann
The linear coefficient in a partially linear model with confounding variables can be estimated using double machine learning (DML).
Methodology Statistics Theory Statistics Theory
1 code implementation • 31 Aug 2021 • Corinne Emmenegger, Peter Bühlmann
Traditionally, spline or kernel approaches in combination with parametric estimation are used to infer the linear coefficient (fixed effects) in a partially linear mixed-effects model for repeated measurements.
1 code implementation • 18 Jul 2023 • Xinwei Shen, Peter Bühlmann, Armeen Taeb
In a linear setting, we prove that DRIG yields predictions that are robust among a data-dependent class of distribution shifts.
1 code implementation • 18 Sep 2023 • Alexander Henzi, Xinwei Shen, Michael Law, Peter Bühlmann
In recent years, there has been a growing interest in statistical methods that exhibit robust performance under distribution changes between training and test data.
no code implementations • 7 Aug 2015 • Christopher Nowzohour, Marloes H. Maathuis, Robin J. Evans, Peter Bühlmann
We consider the problem of structure learning for bow-free acyclic path diagrams (BAPs).
no code implementations • 1 Mar 2016 • Niklas Pfister, Peter Bühlmann, Bernhard Schölkopf, Jonas Peters
Based on an empirical estimate of dHSIC, we define three different non-parametric hypothesis tests: a permutation test, a bootstrap test and a test based on a Gamma approximation.
no code implementations • 25 Nov 2013 • Christopher Nowzohour, Peter Bühlmann
Given data sampled from a number of variables, one is often interested in the underlying causal relationships in the form of a directed acyclic graph.
no code implementations • 6 Oct 2013 • Peter Bühlmann, Jonas Peters, Jan Ernest
We develop estimation for potentially high-dimensional additive structural equation models.
no code implementations • 14 Nov 2013 • Po-Ling Loh, Peter Bühlmann
We establish a new framework for statistical estimation of directed acyclic graphs (DAGs) when data are generated from a linear, possibly non-Gaussian structural equation model.
no code implementations • 11 May 2012 • Jonas Peters, Peter Bühlmann
In this work, we prove full identifiability if all noise variables have the same variances: the directed acyclic graph can be recovered from the joint Gaussian distribution.
no code implementations • 6 Jan 2015 • Jonas Peters, Peter Bühlmann, Nicolai Meinshausen
In contrast, predictions from a non-causal model can potentially be very wrong if we actively intervene on variables.
Methodology
no code implementations • 29 May 2020 • Domagoj Ćevid, Loris Michel, Jeffrey Näf, Nicolai Meinshausen, Peter Bühlmann
Random Forest (Breiman, 2001) is a successful and widely used regression and classification algorithm.
no code implementations • 14 Aug 2020 • Peter Bühlmann, Domagoj Ćevid
We review some recent work on removing hidden confounding and causal regularization from a unified viewpoint.
no code implementations • 23 Jun 2020 • Solt Kovács, Housen Li, Peter Bühlmann
In this discussion, we compare the choice of seeded intervals and that of random intervals for change point segmentation from practical, statistical and computational perspectives.
Methodology Computation
no code implementations • 20 Oct 2020 • Solt Kovács, Housen Li, Lorenz Haubner, Axel Munk, Peter Bühlmann
Change point estimation is often formulated as a search for the maximum of a gain function describing improved fits when segmenting the data.
no code implementations • 6 Jan 2021 • Solt Kovács, Tobias Ruckstuhl, Helena Obrist, Peter Bühlmann
We consider estimation of undirected Gaussian graphical models and inverse covariances in high-dimensional scenarios by penalizing the corresponding precision matrix.
Methodology Computation
no code implementations • 12 Jul 2021 • Michael Moor, Nicolas Bennet, Drago Plecko, Max Horn, Bastian Rieck, Nicolai Meinshausen, Peter Bühlmann, Karsten Borgwardt
Here, we developed and validated a machine learning (ML) system for the prediction of sepsis in the ICU.
no code implementations • 29 Nov 2021 • Mona Azadkia, Armeen Taeb, Peter Bühlmann
DAG-FOCI outputs the set of parents of a response variable of interest.
1 code implementation • 24 Mar 2022 • Zijian Guo, Mengchu Zheng, Peter Bühlmann
The success of TSCI requires the instrumental variable's effect on treatment to differ from its violation form.
1 code implementation • 29 Jun 2022 • Corinne Emmenegger, Meta-Lina Spohn, Timon Elmer, Peter Bühlmann
Causal inference methods for treatment effect estimation usually assume independent units.
no code implementations • 11 Feb 2023 • Jeffrey Näf, Corinne Emmenegger, Peter Bühlmann, Nicolai Meinshausen
The Distributional Random Forest (DRF) is a recently introduced Random Forest algorithm to estimate multivariate conditional distributions.
no code implementations • 2 Apr 2023 • David Carl, Corinne Emmenegger, Peter Bühlmann, Zijian Guo
TSCI implements a two-stage algorithm.
no code implementations • 5 Sep 2023 • Zhenyu Wang, Peter Bühlmann, Zijian Guo
Classical machine learning methods may lead to poor prediction performance when the target distribution differs from the source populations.
no code implementations • 25 Oct 2023 • Christoph Schultheiss, Peter Bühlmann
We propose a method to detect model misspecifications in nonlinear causal additive and potentially heteroscedastic noise models.
1 code implementation • 15 Feb 2024 • Niklas Pfister, Peter Bühlmann
In this work, we extend the nonparametric statistical model to explicitly allow for extrapolation and introduce a class of extrapolation assumptions that can be combined with existing inference techniques to draw extrapolation-aware conclusions.